diff_of_means ratio_of_sd monthly_amplitude_ratio_of_means sign_correlation qqplot_mae acf_mae extremogram_mae
xgboost.mri_esm2_0.ssp245 1.50% 0.900 0.848 0.513 4.417 0.131 0.031
xgboost.mri_esm2_0.ssp434 -2.67% 0.910 0.843 0.504 3.645 0.144 0.043
xgboost.mri_esm2_0.ssp370 -5.73% 0.912 0.842 0.501 3.491 0.126 0.031
xgboost.cesm2.ssp245 -18.64% 0.822 0.806 0.516 9.171 0.052 0.022
xgboost.cesm2.ssp370 -21.16% 0.802 0.782 0.510 10.409 0.056 0.019
xgboost.cesm2.ssp585 -21.50% 0.805 0.796 0.510 10.576 0.064 0.018
cnn.mri_esm2_0.ssp245 -32.08% 1.795 1.301 0.512 20.142 0.473 0.081
xgboost.ec_earth3.ssp434 -35.78% 0.793 0.753 0.497 17.602 0.143 0.034
cnn.mri_esm2_0.ssp434 -36.44% 1.838 1.345 0.503 21.970 0.468 0.099
cnn.cesm2.ssp370 -39.31% 1.890 1.366 0.497 23.553 0.453 0.094
cnn.ec_earth3.ssp434 -39.53% 1.879 1.307 0.503 23.277 0.505 0.124
cnn.mri_esm2_0.ssp370 -41.45% 1.889 1.435 0.486 25.263 0.458 0.053
cnn.cesm2.ssp245 -41.65% 1.881 1.282 0.509 23.233 0.509 0.104
cnn.cesm2.ssp585 -54.24% 1.964 1.267 0.498 29.157 0.527 0.108
nv.mri_esm2_0.ssp245 -60.25% 0.754 0.785 0.503 29.644 0.148 0.021
nv.mri_esm2_0.ssp434 -63.73% 0.773 0.800 0.515 31.354 0.163 0.021
nv.mri_esm2_0.ssp370 -70.04% 0.763 0.823 0.484 34.460 0.105 0.020
nv.cesm2.ssp245 -72.90% 0.780 0.824 0.516 35.866 0.073 0.023
nv.cesm2.ssp370 -76.42% 0.794 0.823 0.521 37.598 0.081 0.029
nv.cesm2.ssp585 -76.44% 0.781 0.841 0.518 37.611 0.076 0.025
nv.ec_earth3.ssp434 -85.80% 0.755 0.796 0.492 42.215 0.094 0.020

Time series of the first days

Distribution of daily values by month

QQ Plot

Distribution of the undownscaled value on days with estimated extremes values.

On the x-axis we have the daily mean (standardized). It says Undownscaled value, but is the daily mean after the downscaling. A good idea is to plot the original undownscaled value.

The purpose of this plot is to illustrate the distribution of P(undownscaled value | we predicted an extreme). This is useful because it reveals how much information we can recover concerning extreme events. If the distribution is skewed to the right, it suggests that we’re predicting extreme values only when extreme values have already occurred. Conversely, if the lower tail of the distribution resembles the reanalysis data, it indicates that we can capture short-duration extremes (e.g., brief periods of heavy rainfall, such as an intense downpour lasting an hour before stopping).

Autocorrelogram

Extremogram